Train Your Own Models

One of the most important things to practise is creating your own models over and over. Don't just use other people's models out of the box without doing anything, because something or somebody else's model is not your model. It means that you can train a model to solve the problem that you want to solve with the performance that you need. And that in the learning process, it's only through training and studying lots of models that you will get this kind of sense oft what the process looks like. What does it look like when it's going well? How do you dbug things that aren't working? You're not given that by normal coding.